Spaces:

Dovakiins
/

qwerrwe

Build error

App Files Files Community

qwerrwe / src /axolotl /prompt_strategies

100 contributors

History: 46 commits

MilesQLi

Update to adapt to sharegpt datasets with "assistant" rather than "gp… (#774)

0800885 unverified 11 months ago

__init__.py

926 Bytes

support custom field for completion from yml (#580) about 1 year ago
alpaca_chat.py

3.67 kB

fixes for alpaca w chatml, and don't include attention_mask w mistral for flash attention (#728) 11 months ago
alpaca_instruct.py

638 Bytes

skip the system prompt about 1 year ago
alpaca_w_system.py

5.49 kB

support user defined prompters, pretokenized datasets in config, local parquet, local arrow files (#348) about 1 year ago
completion.py

2.54 kB

split completion text to sequence_len (#616) about 1 year ago
context_qa.py

3.16 kB

add support for defined train split (#654) 12 months ago
creative_acr.py

6.53 kB

Apply isort then black over 1 year ago
llama2_chat.py

8.14 kB

Fix(message): Improve error message for bad format (#365) about 1 year ago
metharme.py

2.41 kB

feat: add Metharme prompt strategy (#446) about 1 year ago
orcamini.py

1.49 kB

Added Orca Mini prompt strategy (#263) about 1 year ago
pygmalion.py

3.57 kB

Fixed pre-commit problems, fixed small bug in logging_config to handle LOG_LEVEL env var about 1 year ago
sharegpt.py
3.55 kB

Update to adapt to sharegpt datasets with "assistant" rather than "gp… (#774) 11 months ago
sharegpt_jokes.py

981 Bytes

use fastchat conversations template (#578) 12 months ago
user_defined.py

2.63 kB

support user defined prompters, pretokenized datasets in config, local parquet, local arrow files (#348) about 1 year ago